US Patent Application 17804508. PHONEME-BASED TEXT TRANSCRIPTION SEARCHING simplified abstract
Contents
PHONEME-BASED TEXT TRANSCRIPTION SEARCHING
Organization Name
Microsoft Technology Licensing, LLC==Inventor(s)==
[[Category:Yuchen Li of Redmond WA (US)]]
PHONEME-BASED TEXT TRANSCRIPTION SEARCHING - A simplified explanation of the abstract
This abstract first appeared for US patent application 17804508 titled 'PHONEME-BASED TEXT TRANSCRIPTION SEARCHING
Simplified Explanation
- This patent application describes a computer-implemented method for searching and aligning text transcriptions with specified spellings in audio sessions. - The method involves receiving a search query with a specified spelling and generating a sequence of search phonemes corresponding to that spelling. - A sequence of transcript phonemes is also generated from the text transcription. - A search alignment is then generated, aligning the sequence of search phonemes with a fragment of transcript phonemes. - If the search alignment has a quality score exceeding a threshold, it is determined that the fragment of transcript phonemes and the associated portion of the text transcription resulted from an utterance of the specified spelling in the audio session. - A search result is output indicating that the fragment of transcript phonemes and the associated portion of the text transcription are determined to have resulted from the utterance.
- Simplified explanation:
- This patent application describes a method for searching and aligning text transcriptions with specified spellings in audio sessions. - The method involves generating phoneme sequences for the search query and the text transcription. - A search alignment is created between the search phonemes and a fragment of transcript phonemes. - If the alignment meets a quality score threshold, it is determined that the transcript fragment and associated text resulted from the specified spelling in the audio session. - A search result is then provided indicating this determination.
Original Abstract Submitted
A computer-implemented method is disclosed. A search query of a text transcription is received. The search query includes a word or words having a specified spelling. A sequence of search phonemes corresponding to the specified spelling is generated. A sequence of transcript phonemes corresponding to the text transcription is generated from the text transcription. A search alignment in which the sequence of search phonemes is aligned to a transcript phoneme fragment is generated. Based at least on the search alignment having a quality score exceeding a quality score threshold, the transcript phoneme fragment and an associated portion of the text transcription is determined to result from an utterance of the specified spelling in an audio session corresponding to the text transcription. A search result indicating that the transcript phoneme fragment and the associated portion of the text transcription is determined to have resulted from the utterance is output.
